Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Post-work week API feedback

8 views
Skip to first unread message

David Rajchenbach-Teller

unread,
Mar 30, 2016, 11:53:19 AM3/30/16
to dev-proj...@lists.mozilla.org
So, the workweek is over and everybody has finally had time to hurt
themselves with the various APIs – including both the Adapter API, the
Taxonomy API and the REST API.

So, time has come for feedback.

* Are there bad choices that are going to hurt us?
* What is missing that we should implement soon?
* What is annoying that we should improve soon?
* Anything else?

Cheers,
David

Fabrice Desré

unread,
Mar 30, 2016, 12:20:06 PM3/30/16
to dev-proj...@lists.mozilla.org
I think we need to list for each adapter, which getter/setter was
implemented and if we had to workaround limitations of the current
system. For instance, I'm pretty sure that exposing CRUD-like systems or
very dynamic collections of channels is not very well catered for.
Having getter parameters would help a lot.

--
Fabrice Desré
Connected Devices
Mozilla Corporation

David Rajchenbach-Teller

unread,
Mar 30, 2016, 4:43:28 PM3/30/16
to dev-proj...@lists.mozilla.org
I have opened an issue on the use of the Adapter API for CRUD:
https://github.com/fxbox/taxonomy/issues/53

Cheers,
David

David Rajchenbach-Teller

unread,
Mar 30, 2016, 6:57:17 PM3/30/16
to dev-proj...@lists.mozilla.org
An issue I have: Value::Binary sucks for actual binary data. How can we
improve this? What's the best way to send binary data to a client or to
receive Binary data from the client?

Opened issue here: https://github.com/fxbox/taxonomy/issues/37

Cheers,
David

On 30/03/16 18:19, Fabrice Desré wrote:
> On 03/30/2016 08:53 AM, David Rajchenbach-Teller wrote:
>> So, the workweek is over and everybody has finally had time to hurt
>> themselves with the various APIs – including both the Adapter API, the
>> Taxonomy API and the REST API.
>>
>> So, time has come for feedback.
>>
>> * Are there bad choices that are going to hurt us?
>> * What is missing that we should implement soon?
>> * What is annoying that we should improve soon?
>> * Anything else?
>

Marcus Cavanaugh

unread,
Mar 31, 2016, 9:03:30 PM3/31/16
to David Rajchenbach-Teller, dev-proj...@lists.mozilla.org
One of the things that was particularly tedious about implementing Thinkerbell was interacting with the typed nature of { ChannelType, Type, and Value }. I think we can find a solution that provides type-safety without forcing us to centrally define all types in one place. I've posted the proposal here:


0 new messages